Papers
Topics
Authors
Recent
Search
2000 character limit reached

Transformer Encoder Satisfiability: Complexity and Impact on Formal Reasoning

Published 28 May 2024 in cs.LO, cs.AI, cs.CC, and cs.LG | (2405.18548v3)

Abstract: We analyse the complexity of the satisfiability problem, or similarly feasibility problem, (trSAT) for transformer encoders (TE), which naturally occurs in formal verification or interpretation, collectively referred to as formal reasoning. We find that trSAT is undecidable when considering TE as they are commonly studied in the expressiveness community. Furthermore, we identify practical scenarios where trSAT is decidable and establish corresponding complexity bounds. Beyond trivial cases, we find that quantized TE, those restricted by fixed-width arithmetic, lead to the decidability of trSAT due to their limited attention capabilities. However, the problem remains difficult, as we establish scenarios where trSAT is NEXPTIME-hard and others where it is solvable in NEXPTIME for quantized TE. To complement our complexity results, we place our findings and their implications in the broader context of formal reasoning.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (31)
  1. An SMT theory of fixed-point arithmetic. In N. Peltier and V. Sofronie-Stokkermans, editors, Automated Reasoning - 10th International Joint Conference, IJCAR 2020, Paris, France, July 1-4, 2020, Proceedings, Part I, volume 12166 of Lecture Notes in Computer Science, pages 13–31. Springer, 2020.
  2. R. Berger. The undecidability of the domino problem. Mem. Amer. Math. Soc., 66:72, 1966.
  3. On the computational power of transformers and its implications in sequence modeling. In R. Fernández and T. Linzen, editors, Proceedings of the 24th Conference on Computational Natural Language Learning, CoNLL 2020, Online, November 19-20, 2020, pages 455–475. Association for Computational Linguistics, 2020.
  4. Fast and precise certification of transformers. In Proceedings of the 42nd ACM SIGPLAN International Conference on Programming Language Design and Implementation, PLDI 2021, pages 466–481. Association for Computing Machinery, 2021.
  5. Understanding and overcoming the challenges of efficient transformer quantization. In M. Moens, X. Huang, L. Specia, and S. W. Yih, editors, Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 7-11 November, 2021, pages 7947–7969. Association for Computational Linguistics, 2021.
  6. Language models are few-shot learners. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020.
  7. Tighter bounds on the expressivity of transformer encoders. In A. Krause, E. Brunskill, K. Cho, B. Engelhardt, S. Sabato, and J. Scarlett, editors, International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of Machine Learning Research, pages 5544–5562. PMLR, 2023.
  8. Rigorous roundoff error analysis of probabilistic floating-point computations. In A. Silva and K. R. M. Leino, editors, Computer Aided Verification - 33rd International Conference, CAV 2021, Virtual Event, July 20-23, 2021, Proceedings, Part II, volume 12760 of Lecture Notes in Computer Science, pages 626–650. Springer, 2021.
  9. Temporal Logics in Computer Science. Cambridge Tracts in Theoretical Computer Science. Cambridge University Press, 2016.
  10. BERT: pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2-7, 2019, Volume 1 (Long and Short Papers), pages 4171–4186. Association for Computational Linguistics, 2019.
  11. Towards robustness against natural language word substitutions. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net, 2021.
  12. Position information in transformers: An overview. Comput. Linguistics, 48(3):733–763, 2022.
  13. M. Hahn. Theoretical limitations of self-attention in neural sequence models. Trans. Assoc. Comput. Linguistics, 8:156–171, 2020.
  14. Formal language recognition by hard attention transformers: Perspectives from circuit complexity. Trans. Assoc. Comput. Linguistics, 10:800–810, 2022.
  15. On the robustness of self-attentive models. In A. Korhonen, D. R. Traum, and L. Màrquez, editors, Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers, pages 1520–1529. Association for Computational Linguistics, 2019.
  16. A survey of safety and trustworthiness of large language models through the lens of verification and validation. CoRR, abs/2305.11391, 2023.
  17. J. Marques-Silva and A. Ignatiev. Delivering trustworthy AI through formal XAI. In Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Fourth Conference on Innovative Applications of Artificial Intelligence, IAAI 2022, The Twelveth Symposium on Educational Advances in Artificial Intelligence, EAAI 2022 Virtual Event, February 22 - March 1, 2022, pages 12342–12350. AAAI Press, 2022.
  18. W. Merrill and A. Sabharwal. A logic for expressing log-precision transformers. In A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine, editors, Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023, 2023.
  19. W. Merrill and A. Sabharwal. The parallelism tradeoff: Limitations of log-precision transformers. Transactions of the Association for Computational Linguistics, 11:531–545, 2023.
  20. Saturated transformers are constant-depth threshold circuits. Trans. Assoc. Comput. Linguistics, 10:843–856, 2022.
  21. OpenAI. Gpt-4 technical report, 2023.
  22. A survey of the usages of deep learning for natural language processing. IEEE Trans. Neural Networks Learn. Syst., 32(2):604–624, 2021.
  23. Attention is turing-complete. J. Mach. Learn. Res., 22:75:1–75:35, 2021.
  24. M. Sälzer and M. Lange. Fundamental limits in formal verification of message-passing neural networks. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net, 2023.
  25. Robustness verification for transformers. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net, 2020.
  26. What Formal Languages Can Transformers Express? A Survey. Transactions of the Association for Computational Linguistics, 12:543–561, 05 2024.
  27. P. van Emde Boas. The convenience of tilings. In A. Sorbi, editor, Complexity, Logic, and Recursion Theory, volume 187 of Lecture notes in pure and applied mathematics, pages 331–363. Marcel Dekker, Inc., 1997.
  28. Attention is all you need. In I. Guyon, U. von Luxburg, S. Bengio, H. M. Wallach, R. Fergus, S. V. N. Vishwanathan, and R. Garnett, editors, Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA, pages 5998–6008, 2017.
  29. Thinking like transformers. In M. Meila and T. Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event, volume 139 of Proceedings of Machine Learning Research, pages 11080–11090. PMLR, 2021.
  30. A review of recurrent neural networks: LSTM cells and network architectures. Neural Comput., 31(7):1235–1270, 2019.
  31. Explainability for large language models: A survey. ACM Trans. Intell. Syst. Technol., 15(2):20:1–20:38, 2024.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up

Tweets

Sign up for free to view the 1 tweet with 0 likes about this paper.

Sign Up for Free